This article may be perceived as offensive
Only when people cause harm to others should we intervene. But it is not that simple. What about, for example, social harm, offence, and the undermining of morals?
Intervention should only occur when people cause harm to others. However, it is not that simple. What about social harm, offences, and the undermining of societal morals?
We are going to explore the concept of harm to others. In these cases, intervention is clearly necessary. Anyone who attempts to kill someone else or steals someone’s property is crossing a line. They must be stopped and are often punished. In short, coercion is required. This contrasts with harm to oneself, where coercion is only justifiable in exceptional cases, as discussed in the previous article.
Yet things are often not that simple. We are going to look at the more difficult cases. Harm to oneself and harm to others frequently go hand in hand. If you do something foolish, others are often affected too. This is perfect ammunition for those who favour coercion.
But what actually constitutes harm? Scholars do not all agree on this. Do hurt feelings count as harm? You might say no, but this issue is more nuanced than that. And what if there are no identifiable victims, but society as a whole suffers? Does that still count as harm?
Collateral damage
The previous two articles addressed paternalism. Should we intervene if someone gets themselves into trouble? My conclusions were as follows:
Protection of human dignity is not a valid motive for intervention.
Persuasion is the appropriate response when someone is in danger of metaphorically shooting themselves in the foot.
Coercion is only justifiable if there is a threat of harm to others, the person in question is not in their right mind, or communication is impossible (for example, if there is not enough time).
If there is a threat of serious harm to someone, temporary coercion in the form of a time-out is required. This temporary coercion is solely to ensure the individual can make a conscious, free choice and is fully aware of the risks, consequences, and alternatives.
These conclusions owe much to John Stuart Mill’s harm principle. But we face a problem: when we harm ourselves, others usually suffer as well. There are many examples. You neglect your hygiene, and others are bothered by the resulting odour. You get drunk in the pub and then look for a fight because of your poor judgement. Your employer insists on you wearing safety shoes in the factory because, if something heavy falls on your foot, it affects not only you but also your employer. If you fall ill, your employer might be held liable and have to answer to the authorities. In fact, it is difficult to think of examples where behaviour only harms oneself without impacting others.
But there are not just social costs to consider. If you become ill or die because of your own negligence, those close to you may experience deep sorrow. In an attempt to prevent this, they might pressure you to change your ways.
Mill had no issue with this. He believed persuasion was acceptable but coercion was not. As a general rule, sadness is not grounds for coercion, but social costs might be. Let us focus on these social costs.
Mill acknowledged that almost every act of stupidity reflects on others. However, in his view, not all cases justify paternalism:
With regard to the merely contingent, or as it may be called constructive, injury which a person causes to society by conduct which neither violates any specific duty to the public, nor occasions perceptible hurt to any assignable individual except himself; the inconvenience is one which society can afford to bear for the sake of the greater good of human freedom.
— John Stuart Mill, On Liberty (1859)
Mill argued that paternalism should not be considered when individual behaviour only has indirect social effects, does not violate any specific duty (such as tax obligations), and does not cause demonstrable harm to any individual. In such cases, society must bear the inconvenience as the price of freedom.
Feinberg and public charge
Mill’s criteria are insufficient; the externalities of self-harming behaviour need to be considered more deeply. These external effects, known as ‘public charge’, were explored by the American philosopher Joel Feinberg (1926–2004), regarded as an authority on harm. Feinberg expanded on Mill’s harm principle, formulating several important nuances in his four-volume work The Moral Limits of the Criminal Law. In the first volume, Feinberg outlined some key principles for addressing harm to third parties and, by extension, public charge.
No nitpicking: Feinberg supported the Roman law principle de minimis non curat lex: the law does not concern itself with trifles. For instance, while there are legitimate reasons to intervene if someone threatens to jump off a skyscraper, potential damage to the pavement is not one of them. There is no need for laws requiring people to shower, even if some individuals emit a strong odour.
Probability and severity: The likelihood of harm to third parties matters, particularly in relation to the severity of potential harm. A drunk cyclist is less dangerous than a drunk lorry driver.
Proportionality: There must be a reasonable connection between the seriousness of the public charge and the degree of intervention. For example, the fact that naïve users sometimes unintentionally spread computer viruses does not mean that email should be banned, or that a diploma should be required to access the internet.
Beware of statistical discrimination: Discrimination based on unchangeable human characteristics should be avoided. Even if redheads are statistically more likely to smuggle drugs, this does not justify banning redheads from flying.
Higher interests deserve more protection: Higher interests, such as freedom of expression and the protection of privacy, should be given more weight than lower interests, such as curiosity or gossip. Those who spread plausible but incorrect conspiracy theories, causing harm, may deserve protection, while those who share revenge porn likely do not.
Net increase in freedom: Effective management of public charge should result in more freedom overall. For instance, banning rape may inconvenience those who wish to commit it, but it provides far greater freedom to those no longer living in fear of being raped.
Voluntary victims shouldn’t complain: Voluntary exposure to harm or risk cannot justify coercion. If you participate in paintball, you can’t later complain about being hit by paintballs.
Fair competition: Harm caused by losing in fair competition is not grounds for coercion. A gambler who loses a large sum of money has no right to complain about missing out on the grand prize.
Cumulative hypothetical harms don’t count: Hypothetical cumulative harm, such as potential damage caused if everyone fare-dodged on the train, should not count. The problem only becomes relevant when it happens en masse, and thus does not justify intervention in isolated cases. Feinberg subtly noted that even completely innocent activities can lead to harm if everyone were to do them. If everyone vacuums at the same time, the electricity network will become overloaded. But that is no reason to draw up rules when you are allowed to vacuum: the problem does not occur in reality.
Copycats: The final category of harm to third parties involves copycat behaviour. Feinberg cited the example of the 1978 film The Deer Hunter, which depicted Russian roulette and led to an increase in real-life incidents of the same. However, the link between media and violence is difficult to prove, and Feinberg believed caution is necessary before addressing this kind of harm.
Harm must be compensated for
In a previous article, I proposed two universal moral duties:
1. We must not use people merely as a means to an end.2. We must help those who fall below essential living conditions.I added a third in the previous article:
3. Coercion should not be used on others without sufficient justification.When it comes to harm, I now propose a fourth:
4. When someone causes harm to another, they incur an obligation towards that person.How this obligation is fulfilled depends on the context and may vary culturally, but it could take the form of accountability, reparations, compensation, or other forms of satisfaction. Without delving into extensive evidence, I suspect this principle is universally recognised. It stems from moral impulses related to property, compassion, and retribution.
What is harm?
So, what constitutes harm? Do you have to answer for hurting someone’s feelings? Are you responsible if you offend religious sensibilities, disturb the peace, or undermine social mores? Clearly, we cannot let everyone who takes offence dictate our actions.
In this context, I adopt Feinberg’s definition of harm: wrongful setbacks to a person’s interests. Harm, by definition, is detrimental to interests. But the inclusion of the moral element of wrongfulness is key, as it excludes natural events and accidents. For instance, if I serve you a macaron without knowing you’re allergic to almonds, and you require medical treatment, I’ve harmed your interests. But since neither of us knew about the allergy, Feinberg would argue it doesn’t count as harm.
Culpability
The disadvantage of the element of culpability is that it makes the definition of harm more subjective. What one person considers a serious moral offence, another sees absolutely no harm in. Suppose you accidentally infect your partner, with whom you have an open relationship, with a venereal disease. Some see this as an inevitable accident, a consequence of the agreements you have made with your partner. But another, who categorically rejects open relationships, points out that STD’s between faithful partners, as a rule, do not occur.
Causality
A difficult discussion concerns the question of whether harm must be linked to an event. Let's take the drowning toddler in the paddling pool as an example again. You sit nearby, beer in hand. The toddler falls over, head under water. If you don't do anything, it will drown. Of course you get up and put the child back upright. But the question is: if you do nothing, are you causing harm to the child (and the parents)? Intuitively one says yes, of course, but in the end I think differently.
The child's head was already under water, you had no part in that. Her interests were already harmed before you looked. You are not the one who caused the harm, it was the child itself. Of course you have to help the child out of the water. If you don't, you deserve a severe punishment because you have failed in your moral duty. You are acting culpably. But by not saving the child, you are not the cause of the harm.
If you think otherwise, then you have a problem. Because then you consider yourself (and everyone else) morally obliged to repair and prevent all harm that you encounter, if you are able to do so. Then you are morally deficient if you do not pick up all the plastic rubbish on the street that others have left behind. Then your life will be all about fighting injustice, cleaning up other people's mess, constantly addressing people's shortcomings, in short: you will become an overtired whiner, an insufferable do-gooder. If you don't like that, you will have to look for a limit where your responsibility objectively ends. That becomes difficult if you hold everyone responsible for harm that they themselves have not caused.
Non-material harm
A woman in southern Wales was shocked when she arrived home in July 2024. Someone had clearly been in her house. There was a note: “Don't worry, be happy, eat up and scratch.” Everything had been cleaned up, there was some food and a bottle of wine ready for her, and a bowl of sweets on the table. The floor had been mopped, the garbage bin emptied, the bird feeder filled. The perpetrator turned out to be a 36-year-old homeless man, who was subsequently jailed for 22 months. I think it is a harsh punishment, but at the same time I can imagine the woman's fear.
Although the perpetrator arouses sympathy, the woman's fear counts as harm. Her interest in an undisturbed life in her own home, without prying eyes, and without fear of strangers walking in, was clearly affected. But if that means that those emotions can also be a form of harm, what other emotions can be harm, and when? When do fear, anger, annoyance, and hurt feelings count as harm? After all, a life without fear, anger, annoyance and hurt feelings seems just as important as a life without material harm.
There is harm to third parties that is not so easy to demonstrate: if you do something that you consider absolutely harmless, but which others take strong offence to, harm can occur. Examples of this are also available for the taking. You smoke marijuana in the park. You swim naked in the city pond. You swear in church.
If there are people who suffer psychological problems because of your smoking pot, skinny dipping or swearing, can you be forced to stop? Religious conservatives claim harm based on hurt feelings, distress, or harm to public morals. As the religious supporters react more violently, the supposed harm becomes all the more plausible, so the hurt is sometimes exaggerated. There is also a chance of collective obsession: resistance to supposed evil or injustice becomes the object of group dynamics, often further stimulated by demagogues.
That's why you might be thinking: tut-tut, those hurt feelings shouldn’t be such a big deal. People just have to be able to take a punch. Is that so? We once again consult Feinberg. He argued that harm can also consist of being hurt, how moral outrage can deeply affect people. In his book Offense to Others (1984), he gave the thought experiment of a bus you sit in on the way to work. Feinberg outlined thirty-one situations in which people who with you on the bus exhibit offensive behaviour without damaging you or anyone else. The situations vary considerably. I paraphrase.
Two ladies sit across from you and have an incredibly boring conversation out loud. Someone sits within sight with a swastika on his arm. Next to you is someone sitting who smells like he hasn't washed in months. But it gets worse. A couple starts eating live insects and other disgusting, smelly things. They have to throw up. Nearby, a woman openly changes her sanitary napkins; she throws the used menstrual pad on the floor in front of you. A little further away, two men start giving each other a blowjob. A woman commits bestiality with her dog. Yet another makes a very racist speech out loud.
Feinberg gave the examples to show that everyone has a limit, even if there is no material harm. Hurt, shock, disgust, those kinds of emotions are real and can also be harmful. If you no longer want to leave your house because shocking things are happening on the street, you also suffer harm.
Hurt feelings
We can therefore conclude that non-material harm is real, and we are all vulnerable to it. The challenge is that such harm cannot be precisely measured, nor can it be accurately predicted. People's sensitivities vary, and some may exaggerate their vulnerabilities for their own reasons.
It's tempting, then, to dismiss psychological harm. Why should we take overly sensitive individuals seriously?
Although I’m not entirely convinced by all aspects of Jonathan Haidt’s Moral Foundations Theory, his findings are not without merit. He makes a solid case that moral ideals like freedom, fairness, compassion, loyalty, authority, and purity exist in varying degrees across the world.
Why was the Welsh woman so affected by someone entering her home? Likely, it wasn’t just the fear for her safety. A stranger had touched her belongings without permission, disturbing the intimacy of her space. Similarly, why do we take sexual harassment so seriously? Even when there is no bodily harm, those who experience it feel violated — their sense of honour or purity has been damaged.
If someone were to burn a Dutch flag in front of me, I’d probably laugh. At most, I might find it rude or sad. But we can’t ignore the fact that in many parts of the world, people wouldn’t find this funny at all. While I’m accustomed to disrespectful jokes about members of the Dutch royal family, trying that in Thailand would not go over well. Many Thais do not tolerate jokes about their king, and in fact, it’s illegal.
Closer to home, you can expect strong reactions if you claim there’s nothing wrong with animal testing, downplay climate change, label most refugees as fortune seekers, or people on welfare as job-shy parasites. Friendships can easily fall apart over these opinions.
Is all of this considered harm? A study from 2017 showed that people often have a broader definition of harm than we might assume. In some cultures, there are strict protocols for engaging with ancestors, and not following them is believed to harm the dead. Some Christian parents tell their children that masturbation makes Jesus cry, or that an angel dies every time they do it. Emperor Justinianus I even claimed that homosexuality caused earthquakes. When people are asked why pornography is harmful, they often think of the supposed harm rather than its perceived immorality.
When harm occurs, we tend to quickly view behaviour as immoral. But even in the absence of harm, we often link behaviour we find immoral with harm. Take consensual incest, for example. We instinctively see it as wrong, often because of its abusive potential and the risk of unhealthy offspring. But when procreation isn’t involved and it’s fully consensual, many people still feel it’s categorically wrong. When asked why, most insist that it’s harmful, even when it’s clear that no actual harm has taken place.
The researchers suggest that the combination of exceeding social norms (which feels ‘bad!’) and negative emotions (such as ‘yuck!’ or ‘ugh!’) automatically leads to the assumption that harm is involved. While we may not easily agree on what constitutes harm, it’s widely accepted that harm provides a valid reason for intervention.
Given that harm is so difficult to define, what use is Mill’s harm principle? Mill argued that harm to others justifies coercion. But what happens when people have different concepts of harm? How do we reconcile disagreements about the harmfulness of things like homosexuality, animal suffering, blasphemy, or insults?
Just as with physical harm, the harm principle includes certain conditions when it comes to psychological harm, ensuring we don’t overreact.
Let’s not split hairs.
Here in Sarandë, Albania, where I’m currently staying, the beach is seven floors below the street. There’s a lift in the adjacent apartment complex that I use when I don’t feel like climbing up the stairs. Strictly speaking, I don’t have the right to use it, and perhaps I’m causing some inconvenience to the owners. But the impact is so minor that only a serious nitpicker would take offence. It’s like when someone uses your lighter without asking: no one’s going to make a big deal out of it. We don’t pay attention to trivial matters, and this applies even more to annoyances, fears, anger, minor aches, and pains: things that most people forget about an hour later. It also applies to dissatisfaction when things don’t go your way. We’re talking about serious suffering, real harm; those are the things people must be protected from.
Being sensitive doesn’t give you a free pass
You shouldn’t ride a rollercoaster if you have a heart condition, and you shouldn’t deliberately expose yourself to harm. As a Puritan, you shouldn’t go searching for filth on the internet only to complain about it afterwards. People who are mentally vulnerable also have a responsibility to protect themselves. For example, if you’ve experienced war trauma, it’s better to find refuge if a fireworks display is happening nearby, rather than trying to have the show cancelled. The same goes for any heightened sensitivities. If exposed body parts offend you, it’s probably best not to visit the beach, even though you have every right to be there.
Disagreement is fine
The harm principle only applies to cases of culpable disadvantage. You cannot blame someone for normal, natural behaviour, such as homosexuality, even if others take offence and feel harmed by it. Differences in customs, opinions, worldviews, religious beliefs, or moral systems are not inherently problematic. If these differences bother you, perhaps the issue lies with you. It’s natural to have differing opinions, to hold one another accountable for behaviour, and even to argue — this is normal, natural human behaviour. As long as things don’t escalate beyond control, there’s nothing wrong, and there’s certainly no reason to resort to coercion.
In short, coercion is only justified in cases of culpable psychological harm. This means harm caused by deliberately targeting others or by failing to take reasonable precautions. That brings us to:
The precautionary principle
Freedom of expression is widely supported, but where should we draw the line? When should expressing an offensive but non-violent statement be punished? According to Feinberg, this depends on several factors. Firstly, how shocking is the statement? The social value of the statement also matters: there should be more tolerance for offensive political or religious opinions than for statements made purely to provoke or grab attention.
The precautionary principle relates to the extent that people are involuntarily exposed to opinions, and the scale on which this occurs. You don’t have to read a book, watch a film, or attend a meeting. Especially when there is a warning in advance that the content might not be suitable for sensitive viewers. Different boundaries apply when two friends are having a conversation on a park bench compared to someone using a megaphone in a shopping centre. One requires effort to hear, while the other confronts you unexpectedly. The limits may also differ between public demonstrations and private lectures. If audiences at a lecture or cinema screening, or the readers of a book, are warned in advance about potentially offensive content, then freedom of expression should, in principle, be unlimited. Freedom of speech is not at risk; you can say the most shocking things, but you have a responsibility to make sure your audience knows what to expect beforehand.
Take the case of the Muhammed cartoons. Many Muslims found these depictions offensive. Danish and French cartoonists chose to publish images of Muhammed, knowing they would cause offence. The result was horrendous violent reactions. Should those cartoons have been banned? Certainly not. But they shouldn’t be on the cover. The cover should have carried a clear warning: “Possibly offensive to Muslims.” If someone ignored that warning and read the paper, they shouldn’t later complain about being offended.
Social harm
We have arrived at another form of harm to third parties that could constitute grounds for paternalism, namely harm to social trust, morals, public order, public morality or social capital.
In the 1960s there was a lot of debate about sexual morality in Britain. Much was about the legalisation of homosexuality. Another controversial case was the 1962 criminal trial against a certain Shaw. Shaw produced a magazine entitled Ladies Directory, which listed the names and telephone numbers of prostitutes, often accompanied by a nude photograph and listing their specialties. The pervert who created the magazine was convicted of corrupting public morals. The people had to be protected from such immoralities, even if the ladies in question were voluntarily included in the magazine and the owners had voluntarily purchased the magazine and they — one assumes — knew what they would find in the booklet. It gave rise to heated debates. The liberal side of the matter was fervently advocated by the British legal philosopher H.L.A. Hart (1907–1992), including in his book Law, liberty, and morality (1963).
The argument of social harm is mainly found in religious-conservative circles. Conservative legal philosophers such as John Finnis and Robert P. George, for example, are against pornography. Pornography undermines marriage and thus public morality, they say. The harm consists of harm to character and harm to good institutions. Adultery and loose morals can lead to an increase in venereal diseases. Homosexuality, the use of contraceptives and the breakdown of marriage can lead to an ageing population. It is said that if you show a film that glorifies violence or immorality, you undermine public morality and cause social harm. Many conservative thinkers — and don't discount the churches — can point out how modern values can be harmful.
Social harm can occur, for example, if people trust each other less, if people treat each other more aggressively or if people start to feel less responsible for the common good. Let us take an extreme example to see whether public morality is something of value that can be damaged. Imagine a society where no one trusts each other anymore, where violence is the only way conflicts are resolved, where a human life has no value. It's every man for himself. We are familiar with those types of societies: they are countries with dire poverty, often a crippling civil war between warlords, and a powerless, thoroughly corrupt government. They are generally the poorest countries in the world with welfare levels at rock bottom. Public morality is valuable and can indeed be damaged.
However, there are plenty of arguments against this view. Let me explain a few of them.
Public interest is not a basis for coercion
The most radical position, typically found among libertarians, is that social harm is not a valid reason for coercion. Libertarians generally support Mill’s harm principle, but only when the harm can be traced back to individuals.
Who is society? There is no such thing! There are individual men and women and there are families and no government can do anything except through people and people look to themselves first.
— Margaret Thatcher, No such thing as society, interview with Woman’s Own (1987)
The next article in this series will explore when coercion should be used in the public interest, so I’ll leave that discussion for now. But let me offer this thought: imagine I release a successful film that promotes the idea that lying is acceptable. Or I tell people that the idea of keeping promises is harmful and should be abandoned. Or I hold large gatherings where participants are trained to disregard compassion. The harm I cause could be substantial, even though no single individual might be directly affected.
Less freedom is also a form of social harm
I’ve already mentioned Hart’s argument. He proposed the following scenario: let’s say there is a moral sexual code endorsed by the majority, and people are deeply offended by something that consenting adults do behind closed doors. Now, let’s assume that punishment is an effective deterrent, and without it, people’s morals would shift in the wrong direction. Even in this case, Hart argued, criminalisation is not justified. Why? Because sexual morality must be seen in the broader context of societal morality. That morality includes respect for individual freedom, protection of privacy, the right to safety, and protection against deliberate suffering. Tolerance is also part of our moral framework. Punishing people for violating our value system is incompatible with the very tolerance we claim to uphold. If we want to protect society from harmful influences, we must also recognise that intervention may conflict with other important values, such as freedom and tolerance.
In contrast, British judge Patrick Devlin took a different view in his debate with Hart. Devlin didn’t argue that sodomy was harmful in itself, but that it was so controversial that its legal acceptance would harm society. To better understand Devlin’s point, we can use a different example, since British morals in the 1950s and 60s are now outdated.
Imagine if nudism were suddenly allowed everywhere, and millions of people chose to embrace it. Old and young, slim and heavy, attractive and not-so-attractive — everyone would be naked in public. You’d be standing behind a naked person at the checkout, your colleagues would be naked in the office, and the Attorney General would be naked when addressing the press. There’s no immediate material harm in this scenario, but imagine the chaos. Riots might break out, people would clash, and society could become deeply divided between those who support and those who oppose nudism. Is that something we want?
Hart countered by saying that moral values are never fixed. Even if most people find a behaviour offensive today, that doesn’t mean society will still feel the same in 100 years. Furthermore, while a criminal justice approach may be effective in the short term, it doesn’t necessarily support the long-term survival of the “spirit and rules of social morality.” In fact, repression could be counterproductive. Additionally, proponents of “moral policing” often overlook that it’s not always necessary for the state to intervene. Social pressure can be far more effective at discouraging undesirable behaviour.
Weighing the pros and cons
Are monogamous, heterosexual marriages better for public morality than polygamous relationships, same-sex couples, or pornography? Sexual morality evolves and tends to become more liberal over time. However, that doesn’t automatically mean public morality declines or that well-being decreases. The opposite could be true. Consider the countless loveless marriages that persisted due to social pressure. Think of the children raised in households where the parents despised each other, or the deep misery of men and women who couldn’t express their love for someone of the same sex. Then there are the people who faced humiliation or punishment simply because they found love outside of marriage.
In short, if we’re going to bring social harm into the conversation, we need to be fair and carefully weigh the pros and cons. And that’s no easy task.
Intervention is only necessary when immorality enters the public domain
There are immoral actions that harm no one but the person committing them. In line with Mill’s harm principle, this is where the boundary lies. While we may condemn and discourage such behaviour, punishment is a step too far. In principle, what you do alone in your bedroom remains your own business. But what if you find someone who is just as much of a sinner as you are? Then you can do bad things together that no one else will notice. Still nothing much to worry about as long as you keep your secret pleasures inside.
The problem arises when such a group starts actively recruiting new participants or publicly expressing their views. At this point, things can spiral out of control, public morals can be undermined, and the state may feel the need to step in. We see this happening in countries like Russia and many Islamic nations, where public expressions of controversial behaviour are met with state intervention.
There is clearly something wrong with neo-Nazis. Still, I think one should be allowed to hang Nazi rubbish in one’s own living room. And if one has friends with the same hobby, enjoy it indoors. However, the line is crossed when they hold public rallies, create hateful websites, or recruit new members in public spaces.
Let’s summarise
Mill's harm principle is shared widely, including among many Kantians and Aristotelians. In the case of harm to yourself alone, coercion is not legitimate (but information and social pressure are). If harm occurs to others, coercion is required. It is useful as a rule of thumb, but the elaboration is complicated. It is not without reason that Feinberg devoted four heavy books to it. It is often difficult to distinguish between harm to self and harm to others. And it is difficult to determine exactly what harm is. I have adopted the following positions on this:
For harm to self that also affects others, we do not get much further than a list of rules of thumb.
Hurt feelings are real, and can form a basis for coercion. But it is very difficult to determine whether people are putting on a big act. Applying the precautionary principle should resolve that; it can be used to ensure that almost anything may be said and done that does not lead to demonstrable harm, but that others may take offence at.
Social harm exists, and may theoretically form a basis for coercion. But to prove it is not that easy.
For further reading
John Stuart Mill, On liberty (1859)
Patrick Devlin, The enforcement of morals (1959)
H.L.A. Hart, Law, liberty, and morality (1963)
John Finnis, Natural law and natural rights (1980)
Joel Feinberg, The moral limits of the criminal law, volume 1: Harm to others (1984)
Joel Feinberg, The moral limits of the criminal law, volume 2: Offense to others (1985)
Jeffrie Murphy, Legal moralism and liberalism, Arizona Law Review (1995)
Robert P. George, The concept of public morality, The American Journal of Jurisprudence (2000)
Jonathan Haidt, The righteous mind: why good people are divided by politics and religion (2012)
Christopher Tollefsen, Pure perfectionism and the limits of paternalism, in: John Keown, Robert P. George (red.), Reason, morality, and law. The philosophy of John Finnis (2013)
, Toleration (2014)
, Toleration and freedom from harm. Liberalism reconceived (2018)
Chelsea Schein, , The theory of dyadic morality: reinventing moral judgment by redefining harm, Personality and Social Psychology Review (2017)
This was the fourth article in the series on freedom and coercion. There is also an overview of the articles in this series. The next article will be about forced cooperation and coercion in the general interest.